This week, we learned about PCB fabrication by making a FabISP in-circuit programmer. I chose to replicate the board that Valentin Heun made for the class in 2011.

We started by milling PCBs. It took many tries (~3 hours) to get a PCB with all the traces in tact. Some things I learned in the process:

Milling the PCBs

1. The mill really does adjust by 1mm on the z-axis
2. I tried to use the mini sheer to break off the outer parts of the PCB which hadn’t been milled correctly by the 1/32 bit. Ths broke my PCB.
3. I decreased the speed from 4 to 3 on the fab modules. This definitly helped with milling the traces accuratly.
4. I moved the board from one side of the sacrificial layer to the other. One side yielded much better results than the other – possibly because of the evenness of the sacrificial layer.

PCB Mess-Ups

Finally, I milled a PCB that had all the traces in tact

Final Wire Traces

Then, I moved on to stuffing the board, which took another few hours. Some things I learned during this process:

1. The ATTiny44’s legs may look soldered, but sometimes are not
2. Zener diodes have a direction, as does the ATTiny44
3. Placing a dot of solder on one side, dragging the component into place using tweezers, and then applying a little bit more solder to secure the connection seemed to work the best.

Stuffed PCB

Finally, I programmed the PCB, which failed the first few times due to two of the pins of the ATTiny44 not having a good electrical connection to the board. With lots of help from Palash and Juliana (Cherson), I troubleshooted the problem and got the board working.

It programmed! I was so excited that I took a photo of the terminal.

Lastly, here is the final product, with the USB blocker snapped off.

Final FabISP In-Circuit Programmer

Thanks to Palash, Bianca, Guy, and Juliana (Cherston) for help troubleshooting this week.